Subject: Linux-Development Digest #33
From: Digestifier <Linux-Development-Request@senator-bedfellow.MIT.EDU>
To: Linux-Development@senator-bedfellow.MIT.EDU
Reply-To: Linux-Development@senator-bedfellow.MIT.EDU
Date:     Tue, 16 Aug 94 17:13:13 EDT

Linux-Development Digest #33, Volume #2          Tue, 16 Aug 94 17:13:13 EDT

Contents:
  1.1.43 WD8003 broken? (Steven K. Chang)
  Where is irqaction? (Robert Bauer)
  Re: Future of Linux (David Kastrup)
  Re: sbpcd ejects on umount in 1.1.44 (Patrick Doyle)
  1.1.45 NCR 53c810 problems (Larry Augustin)
  Re: Was: [hd] Do you recognize this error message? Now: Error survey (Paul Cadach)
  Re: Can't compile/build kernel 1.1-36 (Brian M Berry)
  Re: IPC in Linux. (Matthias Urlichs)
  Re: Aztech and Glalaxy card compatibility with linux?? (NARDONE@DELPHI.COM)
  Re: Where is irqaction? (Sid Boyce)

----------------------------------------------------------------------------

From: stchang@leland.stanford.edu (Steven K. Chang)
Subject: 1.1.43 WD8003 broken?
Date: 16 Aug 1994 16:13:07 GMT

We were trying to set up a network yesterday, and Linux 1.1.43 failed to
recognize my ethernet card, an SMC Elite Ultra based on the WD8003 chip.  

It was an eight bit card.  We think everything was configured correctly
because when we reverted back to 1.1.0, 
it works.

Just letting you all know (if you don't already).  I am a Linux newbie, so
it could have been my fault.

-- 
Steven K. Chang
stchang@leland.stanford.edu

------------------------------

From: rbauer@ecst.csuchico.edu (Robert Bauer)
Subject: Where is irqaction?
Date: 16 Aug 1994 17:01:13 GMT

Hello,

I noticed with kernel 1.1.45 (and possibly earlier releases
which I skipped) that irqaction is no longer listed in ksyms.c.

Since this breaks ftape, I was wondering if there is a reason behind
this omission, or if it was simply forgotten.  Can I just stick it
back in there with no ill consequences?

Thanks,

Robert
-- 
      _____ _    __     | rbauer@ecst.csuchico.edu
 _ __|___  | |_ / _|___ | N7TFZ@KE6LW.#NOCAL.CA.USA.NA
| '_ \  / /| __| ||_  / |------------------------------------------------
| | | |/ / | |_|  _/ /  |
|_| |_/_/   \__|_|/___| |                         "Unix wants to be free"

------------------------------

From: dak@rama.informatik.rwth-aachen.de (David Kastrup)
Subject: Re: Future of Linux
Date: 16 Aug 1994 18:03:32 GMT

byron@gemini.cc.gatech.edu (Byron A Jeff) writes:

>In article <32n4pk$b7d@wumpus.cc.uow.edu.au>,
>Van Dao Mai <mai@wumpus.cc.uow.edu.au> wrote:
>-Guys,
>-   I read a debate on comp.os.linux.misc about some people thinking of
>-Linux as only a toy. I like it very much as it is free and smarter than
>-many commercial systems. It has faults that are common of all UNIXes.

>Yes. That fault seems to be is that Unixes aren't DOS/Windows/Macs.

>-   I suppse one has to be clear of the purpose of a system to see its
>-value. If UNIX is still targeted at education mainly then it is fine as 
>-it is. However if it is targeted at normal users then it is not.

>Because normal users use DOS/Windows/MacOS.
[...]
>-May be this will give us a good set of Linux software that will make it on
>-the same footing as Windows, OS/2 or DOS. The most urgent things to have t
>-for Linux is 
>- (1) a decent object-oriented filemanager for X windows and

>If none of the X file managers available can't do the job then why not
>pick the best one and start working on it. Identify what needs to be
>changed, soliclit some help, and get to work.

>Personally I'm perfectly content with the X utilities menu and a shell.


>-(2) A powerful word processing with GUI interface for Linux. 

>If the EZ editor from the AUIS doesn't do the job then why not start working
>on it. Identify what needs to be changed, solicit some help, and get to work.

>Personally I'm much more effective using LaTeX or Quickscript, ghostview
>and a PostScript Printer.

>- These two would make the system a lot more useable for all people. 

>To be honest it doens't make the system any more useable to me.

>-
>-This will make Linux a home UNIX system for many "better than average"
>-computer user.

>I'm tired of this argument. People just can seem to understand the amount
>of people-hours, management, and money that it takes to develop the kinds
>of software that you're talking about. Each one of the projects you're
>referring to have hundreds of people working on it full-time. The reason
>software companies can do that is because they know they'll get 15 million
>units sold to a extremely loyal customer base. It's a lot easier to work
>for years on a product when you know that even 2% of a 150 million
>installed base will buy it.

>But when you're essentially giving away the product and there's a small
>customer base, it becomes very difficult to justify all that effort.
>Especially when there are tools available to do the job.

>The better way to support all this stuff is to provide the underlaying
>OS services necessary to support such applications. DOSEMU is a champ at
>this while Wine is moving into usefulness.

>The big software companies have spoiled users into having these
>"everything including the kitchen sink" applications that have been developed
>and refined for 10 years or more. If you take a survey of applications
>users you'll find that the 90/10 rule applies in that most (90%) of the
>users use only a small (10%) of the available functions. But because the
>customer is paying they can demand the other 90% of the functions even though
>they are rarely used by any but the most rare user.

>The challenge is to identify the 10% of functions that are most used and
>implement them. Add hooks to that users can add anything they want. 

>Face it Linux is useful and unique because it doesn't do the same things as
>DOS/Windows/MacOS. If that's all you need then stay with DOS/Windows/MacOS.
Stop. Disagreement here. The problem is that more and more people become
dissatisfied with DOS/Windows at least, because they become more and more
a botch of unnecessary compatibility idiocies. What other 32-Bit-System
is on market, which feels it has to support FCBs (low-level file handling
structures for non-directory-tree-structured file systems imitated from
CP/M for compatibility reasons) and other BDOS calls from ancient OSses.
What operating system
(or combination) shoots its foot with memory management the same way
a DOS/Windows combination does? And not only its own foot, but that of
developpers as well. In integrated environments the user is shielded
(apart from slow disk access, idiotic file system restrictions, slow
OS interfaces, memory hassles, configuration wrestling...), but the
developper is not, and the user pays for the battle of the developper
in order to get spared the OS himself.

Now MacOS has lots less of compatibility problems, and so would be
ideal for the user base mentioned. Except that their long-time high-price
policy has narrowed the number of users, and consequently developpers.
Also, with the prevalence of C, Unix system calls (or their library
equivalents) and tools have become rather popular. But this is
somewhat coupled with the command line paradigm (what would Unix
be without pipes?) and this is somewhat neglected in MacOS.

For closed applications, use MacOS, if you need, as it is rather
properly done (apart, maybe, from TCP/IP support, but this is changing,
I believe).

For command line systems with toolbox philosophy, use Linux. Do not
expect free time developpers to make something like *ord/*ordperfect,
if they have emacs and vi. Programmers enjoy programming better
than writing books with a pampering, limited word processing package
with mediocre results.

>Linux has no need to compete with DOS/Windows/MacOS. The user bases of the
>others are so large and so entrenched that trying to make even the smallest
>dent is impossible. All we really have to do is provide a high quality
>product whose worth is easily identifiable to knowledgeable users.
Right. It is just that its interface is very alluring for those still
using the MSDOS dinosaur OSs, as they offer really nothing better apart
from dinosaur based applications.

>I've stopped trying to get people to convert. For most Linux is unnecessary. 
>I'll only push the point when I have some responsibility to supporting the 
>machine.  I'm selfish because I'm no longer interested in supporting
>DOS/Windows/MacOS boxes. There are enough other people/books/magazines
>for those products. I do Linux. If I'm involved, Linux is involved. If
>Linux isn't involved, then I'm not. 
But probably you might be convinced to develop for Unix, if you were
allowed using Linux for development?

>I've brought Linux in its current form to everywhere I operate. People have
>been impressed with it's strengths and not so worried about it's lack of
>applications because of the DOS/Windows/MacOS environments that are
>prevalent for doing the types of applications you describe.
Right. And I'd not worry too much. Linux is one hell of a developping
system, and has a very large potential user base in the private sector.
There will be times when corporate manpower will turn to Linux development
for efficiency's sake. Especially as your programs will compile on
almost every other Posix system.

>Most average computer users have no understanding or need for Linux.

>All you'll end up doing is emasculating the OS we know an love by chaining
>more and more restrictions so that it'll look/feel like DOS/Windows/MacOS.
Stop. The poster complained not about the OS, but about the applications
available.

Of course, I'd prefer custom programs to have file formats I can munge
with awk, if I so desire.
-- 
 David Kastrup        dak@pool.informatik.rwth-aachen.de          
 Tel: +49-241-72419 Fax: +49-241-79502
 Goethestr. 20, D-52064 Aachen

------------------------------

From: wdoyle@hilbert.coe.northeastern.edu (Patrick Doyle)
Subject: Re: sbpcd ejects on umount in 1.1.44
Date: 16 Aug 1994 18:06:53 GMT

In article <32gmvo$8fo@cantua.canterbury.ac.nz> drb@chem.canterbury.ac.nz (Ross Boswell) writes:
>I don't want sbpcd to eject the tray when the device is unmounted --
>it's likely to crunch my fingers!
>At 1.1.41 I edited drivers/block/sbpcd.c to define EJECT (and JUKEBOX)
>to zero, and this vicious behaviour was cured.
>At 1.1.44 it has returned, although the defines remain zero and there
>is no new call to yy_SpinDown introduced into sbpcd.c.  
>
>Any suggestions?

I decided I didn't like this feature either, so I poked around sbpcd.c
and found that the only reference to EJECT was something of the form:

#ifdef EJECT
  yy_SpinDown(...)
#endif

Which means the yy_SpinDown will be called regardless of the value of
EJECT.  I simply placed an "#if 0/#endif" pair around this piece of
code and now the CDROM does not eject automatically.  I found this
feature particularly annoying when I started "workbone", started
playing a CD, and exited workbone.  Upon exit, the CD ejected!  (Thus
making it dificult to continue listening to the lyrical tones of Billy
Joel :-).

------------------------------

From: lma@dayton.Stanford.EDU (Larry Augustin)
Subject: 1.1.45 NCR 53c810 problems
Date: 16 Aug 94 17:09:09 GMT

These are both with the 1.1.45 kernel.

NextStor NCR 53c810 PCI SCSI card, Micronics P54i PCI motherboard:
    scsi0: using io mapped access
    scsi0: using initiator id 7
    scsi0: using edge triggered interrupts
    scsi0: burst length 8
    scsi0: using 40MHz SCSI clock
    scsi0: m_to_n = 0x90, n_to_m = 0x90, n_to_n = 0xb0
    scsi0: testing
    scsi0: test 1 started
    scsi0: tests complete
    scsi0: NCR53c{7,8}xx (rel 3)
    scsi: 1 hosts.
    scsi0: DANGER: command in running list, can not abort
    scsi0: DANGER: command in running list, can not abort
    SCSI host 0 abort() timed out - reseting
    Unable to reset scsi host 0 - probably a scsi bus hang

ASUS SP3G, on-board NCR 53c810:
    system hangs with floppy read light on immediately after detecting
    the 53c810 and determining its io, irq, etc.

BTW, I've sent 2 email messages to Drew Eckhardt offering to loan him
a 53c825 board, but haven't heard any response.  Does anyone know if
he's still working on the drivers?  Alternatively, is there someone
else working on the drivers who could add support if they had a board
to test?

Larry


------------------------------

From: paul@paul.east.alma-ata.su (Paul Cadach)
Subject: Re: Was: [hd] Do you recognize this error message? Now: Error survey
Date: 16 Aug 1994 23:44:23 +0700

Chris (e8ne@jupiter.sun.csd.unb.ca) wrote:
> In article <32mro0$8av@cronkite.ocis.temple.edu>,
> The Answer is 42. <jwiegand@odie.temple.edu> wrote:
> >In article <317fdd$bvd@aurora.engr.LaTech.edu> ramos@engr.latech.edu (Alex Ramos) writes:
> >>Might some hd.c guru tell me a possible cause for the following errors?

> [snip] (HD errorors)

> >[snip]

> >I have developed soft errors on my hdb, and the soft errors result in
> >file/file-system corruption. The problem, as far as I can tell, is
[skipped]
> df results in the following (I am at work, but you should see my point)

> avail  used    free
> 135xxx 119xxx 10xxx    *note: 119+10 != 135

E2FS have some "reserved" blocks (usually about 5% of full partition size).
You can change this value by 'tune2fs'.

> I have rebooted several times, and STILL I get about 6/8 megs MISSING!?!

It's your 5% (I think its reserved for "critical" moments when free space
must be less than zero)!

[skipped]
> >So the big question is: what is the right way to recover from hd errors?
> >I don't have the AT BIOS listings or IDE controller documentation on the
> >register level, so hopefully someone can point me to documentation or
> >give me a hand :-). 

> Please - I'd LOVE to hear about this too :)

Some time ago I looked for AT BIOS's HDD IO operations. It's very
simply - all such as in kernel, but all delays realized on "dead
loops" (i.e. looks like this:
        mov     cx,0FFFFh
L_LOOP:
        in      al,dx
        test    al,80h
        jz      L_OK
        loop    L_LOOP
L_FAIL:
). I not have full HDD documentation too (Flambeaux's TeachHelp 4.0 only).

> >So, if you find filesystem problems on your IDE partition, don't blame
> >Linus or Linux, check your syslog for HD write_intr errors. Getting one

> I believe my problem IS related to the Kernel... *sigh* But I did NOT get 
> any errors in /var/adm/*log about HD errors...

I have many Maxtor 2585-AT hard disks which have ugly Hong-Kong's on-HDD
controllers. Often I have a next situation: disks' speed increases about
up to 2 times of its standard then decreases to normal value. At this
moment the controller generates (read|write)_intr with error code = 0x59 and
status = 0xFF (I not remember, but first value in the error message was
0x59, second - 0xFF), so the HD's driver returns with fault as when I have
a bad block. In this situation many programs does cannot be started,
because they code segment is wrong. I patched it by commented out
"bad_block's" check in the linux/drivers/block/hd.c.

Also, when I using ide-performanse patch with MultModeValue=16 (maximum
for those Maxtor's), I often have a situation like I described before.
But when I changes this value to other value less than 16 all works more
better. I can't understand why its does.


Sorry for my broken english.

With best regards,
Paul.
-- 
Paul Cadach        System Programmer | kADA^ pAWEL   sISTEMNYJ PROGRAMMIST
e-mail:   paul@paul.east.alma-ata.su | FIDO:    2:5083/11.39
phone:    +7-(3232)-66-0538          | TEL.:    +7-(3232)-66-0538
=====================================+====================================

------------------------------

From: berry@bdmserver.bdm.com (Brian M Berry)
Subject: Re: Can't compile/build kernel 1.1-36
Date: 11 Aug 1994 13:42:50 GMT

Sam Gentile (owlmed@mv.mv.com) wrote:
: I obtained the 1.1.35 source and unpacked that. I then patched with the 
: 1.1.36 patch and started the traditional build process. It went a long ways
: and then:
: /usr/src/linux/include/linux/timer.h: In function 'init_timer':
: In file included from psaux.c:28:
: /usr/src/linux/include/linux/timer.h:85:  'NULL' undeclared (first use 
: this function)
: make[2]: *** [psaux.o] Error 1
: make[2]: Leaving directory '/usr/src/linux/drivers/char'

: Can someone please tell me how to fix this error? I am completly stuck 

I encountered the very same problem.  Although a dirty fix, I edited 'timer.h'
and instructed it to include (if I remember this right) 'stdio.h'... check
to make sure this is the correct one---just grep for a "#define NULL" in
/usr/include.

This seems like a pretty silly error---and while I don't generally like
to deviate from "pure" kernel source, adding a single line shouldn't
botch things too badly :-).

In any case, after making this adjustment, 1.1.36 compiled correctly and
has worked just fine for me in the past week (with thorough use).

Good luck!

--
Brian M Berry                 Phone: 703/848-5876 or 800/955-BDM1 ext. 5876
BDM Federal, Inc.             Internet: berry@mcl.bdm.com
1501 BDM Way                  (This message printed with 100 percent
McLean, Va 22102-3204         recycled electrons.)

------------------------------

From: urlichs@smurf.noris.de (Matthias Urlichs)
Subject: Re: IPC in Linux.
Date: 16 Aug 1994 22:15:54 +0200

In comp.os.linux.development, article <32odgs$3p6@venus.mcs.com>,
  macgyver@MCS.COM (MacGyver) writes:
> 
>       passing file descriptors -- ability to pass open file descriptors
>                                       betwen processes.
> 
> there via preexisting functions.  However, I DONT see any support for passing
> file descriptors at all and it SHOULD be there.  Now, either I'm not looking

Nope, it isn't. You can pass the process ID and the file number and tell
the other guy to go and open /proc/PID/fd/NUM. Not the same, of course,
but when in need...

> While I'm on the subject, what about streams ala SVR4?  Things like putmsg()
> don't seem to exist in Linux either...is it planned?  Same questions as
> above...if no one's done it and it's not planned or being worked on, I'd
> like to take a crack at it possibly.
> 
Right now I'm massaging my Streams port (originally done to support an ISDN
driver) to support TTYs and some lookalike of the "clone" driver.
(Actually, TTY support is easy, and so is a clone driver once you hack the
inode support, but marrying the two is not that simple.)

Basically, it's almost, but not quite, entirely unlike Streams.  ;-)
As of now, it's something like a bare-bones SVR2 Streams. This means: no
multiplexing modules, no file descriptor passing, no sendmsg/recvmsg, no
max/min packet sizes. You want a feature, you add it yourself.

I'll publish the package as soon as its TTY support has settled down a bit.
I can't do a successful login through the Streams loopback driver...yet.

-- 
Fleas can be taught nearly anything that congressmen can.
-- 
Matthias Urlichs        \ XLink-POP N|rnberg  | EMail: urlichs@smurf.noris.de
Schleiermacherstra_e 12  \  Unix+Linux+Mac    | Phone: ...please use email.
90491 N|rnberg (Germany)  \   Consulting+Networking+Programming+etc'ing     42

Click <A HREF="http://smurf.noris.de/~urlichs/finger">here</A>.

------------------------------

From: nardone@news.delphi.com (NARDONE@DELPHI.COM)
Crossposted-To: comp.os.linux.help,alt.cd-rom
Subject: Re: Aztech and Glalaxy card compatibility with linux??
Date: 16 Aug 1994 20:59:00 -0000

barker@numbat.cs.rmit.OZ.AU (Gregory James Barker) writes:

>Is there any way to access an Aztech cd-rom drive through
>the Galaxy sound card?

As it was stated to me in an earlier post, Aztech's newer
drives are proprietary and are NOT supported under Linux.

A call to their tech support line revealed that to get any
technical info on the drive would require a call to _Singapore_.

2 Questions--

   is anyone working on a driver for this for Linux?  if so,
   do you need any help?

   if not, does anyone have any info on this drive?


>The CDROM-HOWTO FAQ says that a Galaxy sound card interface is
>supported. How does this need to be done??

This I don't know...



I'm willing to dedicate the time and programming prowess if
I can get my hands on any solid info.  

THanks,
Joe Nardone
nardone@delphi.com

>I have just bought linux on CD-ROM and it would be really nice to
>install Slackware directly onto the linux partition.

>I would appreciate any help.

>Thank you.

>- Greg.

------------------------------

From: szb50@ccc.amdahl.com (Sid Boyce)
Subject: Re: Where is irqaction?
Reply-To: szb50@JUTS.ccc.amdahl.com (Sid Boyce)
Date: Tue, 16 Aug 1994 18:41:18 GMT

   Getting the same problem and also verify_write undefined. Did a
"make mrproper" and still no luck. Used 1.1.39 and 1.1.44 with
gcc2.5.8 and 4.5.26 libs.
   Someone suggested in a post that irqaction line be changed to
something like irq.request .........., that failed miserably.
   The above happens at work. At home, before I tried to switch to my
new IDE drive, I had 1.1.39 and ftape 1.13b working fine, and GUESS where
my complete linux system is - right - ON FTAPE.
Regards
Sid ... G3VBV .... Amdahl(UK) ....

------------------------------


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: Linux-Development-Request@NEWS-DIGESTS.MIT.EDU

You can send mail to the entire list (and comp.os.linux.development) via:

    Internet: Linux-Development@NEWS-DIGESTS.MIT.EDU

Linux may be obtained via one of these FTP sites:
    nic.funet.fi				pub/OS/Linux
    tsx-11.mit.edu				pub/linux
    sunsite.unc.edu				pub/Linux

End of Linux-Development Digest
******************************
